Select manipulation of natural killer (NK) cell activity in vivo will be used to establish the importance of NK cells in relation to: (1)\transplantation resistance against tumors; (2)\primary oncogenesis; and (3)\regulation of normal bone marrow stem cell function. NK-sensitive or -resistant variants will be selected from tumor cell lines using in vivo selection protocols or controlled induction of differentiation in vitro. The underlying basis for this variability of susceptibility to NK lysis will be analyzed. Regulation of NK activity in vivo will be produced using genetic variants with defined variations in innate NK activity after administering select immune sera in vivo. Transplantation of tumor cells will be done correlating the relative NK susceptibility in vitro to the NK levels in vivo of the recipients of the respective tumors. Primary oncogenesis will be carried out using chemical or viral agents in mice with select deficiencies in the immune system. Bone marrow stem cell numbers will be calculated using in vitro colony growth systems. We will select variants of tumor cells according to NK susceptibility using controlled differentiation in vitro of tumor cells following differentiation in parallel to variations in NK susceptibility. Cloning of cells within the same tumor line displaying morphological markers of changed differentiation will be carried out for NK analysis. Natural killer cells represent one cell type with interesting properties making it a candidate for regulation of certain types of malignant and normal cells in vivo. The present study should lead to a deeper understanding of the physiology of the natural killer cells and their potential roles in vivo during primary oncogenesis, as well as possible regulatory function as to bone marrow stem cells. (SR)